TY - JOUR
AU - Hossain,, Delwar
AB - Abstract High-resolution geophysical techniques can be employed as a means of characterizing the lithological changes within materials frequently known to be variable. Ground penetrating radar (GPR) profiling using 50, 100, 200, and 400 MHz antennae and electrical resistivity imaging have been used to investigate high-conductivity United Kingdom Triassic sandstone lithology and moisture content changes. The investigation site is located outside the School of Geography, Earth and Environmental Sciences at the University of Birmingham on a gentle grassy slope. Three GPR and electrical imaging lines were completed over this site. The results of the observations reveal a higher degree of both vertical and lateral heterogeneity of the highly conductive sandstones. The results obtained using these two high-resolution geophysical tools agree reasonably well with each other. These techniques appear to be useful for high resolution and continuous mapping of the subsurface sediments. electrical resistivity imaging, GPR, UK Triassic sandstones, radar signal penetration depth, moisture content, grain size variation 1. Introduction The Triassic sandstones of the Wildmoor Formation constitute an important aquifer in some parts of the United Kingdom (UK), principally the Midlands and northwest England. High-resolution geophysical techniques can be employed as a means of characterizing the unsaturated zone of this formation. The potentially high resolution of ground penetrating radar (GPR) profiling and electrical imaging, and the sensitivity of GPR to contained fluids and electrically conductive materials, make them attractive for investigations of the shallow subsurface, including small-scale structural controls. Electrical imaging surveys have been used in the characterization of Quaternary sediments' architecture (Pellicer and Gibson 2011), delineating fracture systems within the unsaturated zone of chalk (Andrews et al1995), revealing fluid movement through the unsaturated zone (Barker 1996, Arora and Ahmed 2011), and have been used in various geologic, hydrogeologic, environmental and engineering investigations (e.g., Chambers et al2006, Vaudelet et al2011). The GPR methods have various applications, including in studying dunes (Gómez-Ortiz et al2009), mapping the water table, investigating sand and gravel aquifers and stratigraphy (e.g., Davis and Annan 1989, Beres and Haeni 1991, Pipan et al2003), determining the moisture content in soil (Turesson 2006), and mapping contamination plumes (e.g., Davis and Annan 1989), among others. GPR techniques have been used to solve many other practical problems over the past few decades (e.g., Davis and Annan 1989, Turner et al2011). Various possible applications of electrical tomography have been discussed by, e.g., Barker (1996), Giao et al (2008). The basic principles of the GPR method have been described by, e.g., Davis and Annan (1989) and Jol (2009). Fisher et al (1992) and Gerlitz et al (1993) have presented techniques for processing GPR data which utilize the similarities and inherent differences between GPR and seismic data types. The principles of electrical tomography have been described by, e.g., Loke and Barker (1995, 1996), Barker (1996) and RES2DINVx32 version 3.71 (Geotomo Software 2012). The objective of the work presented here has been mainly to carry out studies on the high-conductivity UK Triassic sandstone lithology and to identify moisture content variations with GPR and resistivity profilings. 2. Site description The study site is located outside the School of Geography, Earth and Environmental Sciences, at the University of Birmingham on a gentle grassy south-westerly facing slope in-between two footpaths (figure 1). A number of tensiometers, piezometers, neutron access tubes and an open hole (10 m depth) are situated on the site within a 4 m radius of a 58 m deep borehole. Figure 1. Open in new tabDownload slide Location map of the study site at the University of Birmingham, UK. The GPR and resistivity profiles, and holes and tensiometers, are shown. Figure 1. Open in new tabDownload slide Location map of the study site at the University of Birmingham, UK. The GPR and resistivity profiles, and holes and tensiometers, are shown. The study site is situated on the Triassic sandstones of the Wildmoor Formation. The profile consists of a sandy soil down to about 0.6 m depth, which grades into uncemented sand. The sand becomes more consolidated at approximately 1.8 m, but the sandstone is very poorly cemented all through the profile (Digges La Touche 1998). Occasionally, large cobbles are available in the shallow soil zone. The sandstone bedding here is shallowly dipping. The sand seems to have increased clay content in the depth range 1–3 m. Below 3 m depth, the medium-sized sand is of red–brown colour. At about 7 m depth, there is an interconnected, manganese-lined macro-pore network with pores of about up to 4 mm in diameter. Breaks in the core below 3 m in depth were logged as fractures but could equally represent bedding planes, uncemented horizons or changes in grain size; it may, therefore, be more correct to refer to these as discontinuities. Depth to the water table varies (in time) from 8 to 10 m. Hydrogeological work at this site indicates that porosities range between 4% and 26%, while vertical permeabilities range from just under 0.01 to over 4000 mD (Berry 1996). The unsaturated hydraulic conductivity appears to vary within the range 0.03–1 mm day-1 (Digges La Touche 1998). 3. Data acquisition and processing The application of the electrical and GPR profiling methods is advantageous because of their suitability to cover large areas in a limited time; the former methods usually require limited processing of data. 3.1. Data acquisition The Ramac/GPR system manufactured by Mala Geoscience (MALA 2012) was used for the surveys. The system consists of a transmission antenna (1000 v) (pulse at high amplitude, typically 1000 V), receiving antenna, control unit, power sources, trigger box and hip chain. The acquisition is controlled using a laptop computer which allows the data to be displayed as they are collected. Figure 1 shows the location and layout of the survey. The data consist of common-offset profiles acquired with 50, 100, 200 and 400 MHz antennas. A variety of frequency antennae was chosen, because emphasis was placed on both the depth of penetration and the resolution. Three 2D lines measured at 4.8 m intervals were 39.2 m long. Initial parameter tests were run by recording common-mid-point gathers at 100 MHz, with an initial antenna separation of 1 to 31 m in steps of 1 m, to determine the velocity, to enable depth calculation and to define optimal offset and recording time. The near-surface velocity determined from the direct ground wave is about 0.086 m ns-1 (figure 2). Figure 2. Open in new tabDownload slide A common-mid-point profile used to calculate the near-surface radar-wave velocity at the study site. Air-wave reflections are seen at times around 200 ns. Figure 2. Open in new tabDownload slide A common-mid-point profile used to calculate the near-surface radar-wave velocity at the study site. Air-wave reflections are seen at times around 200 ns. The antenna dipoles were perpendicular to the inline direction. In order to facilitate a comparison of the records, the GPR lines were collected with identical distances between two adjacent points of measurement (0.1 m), and antenna separation (1 m). The number of stacks per trace varied from 16 to 64. The sampling interval varied depending on the antenna frequency. For the electrical imaging surveys described below, a Campus multicore cable imaging system, using a unit electrode spacing of 0.8 m, was employed with 50 electrodes deployed in a Wenner array. For such an array, the maximum possible measurement separation is around 13 m, giving a depth of investigation of approximately 6.5 m. The cables were connected to a switching module in a Campus Geopulse earth resistance meter. All of the resistance measurements were collected in traverse mode automatically under the control of a laptop computer linked to the Geopulse through an RS232 port. 3.2. Data processing The GPR processing and interpretation software Gradix VI (Interpex Ltd 1996) was used to perform the processing of the GPR profiles. The processing steps used to enhance the signals of the GPR profiles included dewowing, declipping, filtering, setting time zero, gain application and trace mixing (Interpex Ltd 1996). The automatic resistivity inversion technique developed by Loke and Barker (1995, 1996) has proved to be markedly successful in eliminating electrode geometry effects, so that the final processed image provides a good representation of the subsurface. This technique is based on the smoothness-constrained least-squares method (de Groot-Hedlin and Constable 1990), and it produces a two-dimensional subsurface model directly from the apparent resistivity pseudosection. The l2-norm inversion method has been used in this case. 4. Field results In order to study the top of the UK Triassic sandstones, electrical resistivity and GPR profilings and moisture content data have been used. The GPR signal penetration depth discussed below is as determined from 50 MHz profiles. In the recorded GPR sections, the resolution of the reflectors is poor, especially with the 50 MHz antennae. The noise, like air-wave reflections and air-wave diffractions, might have been recorded at times greater than about 200 ns, and, therefore, does not have any interference with signals on the records involved. The GPR records for profile 1 (figure 3) reveal an almost parallel, discontinuous at the left side, reflection pattern, the reflectors being consistent with bedded and discontinuous sand and soil, and change in the moisture content. Beres and Haeni (1991) have interpreted parallel reflectors as fine-grained sediments in the glaciated northeast. The combined interpretation of the records obtained using 50, 100, 200 and 400 MHz antennae (figure 3) has revealed a few reflectors in the shallow subsurface as discussed below. Figure 3. Open in new tabDownload slide GPR records and electrical image for profile 1. 50 (a), 100 (b), 200 (c), and 400 MHz (d) GPR records, and electrical image (e). Major lithological changes in the near-subsurface are indicated. Figure 3. Open in new tabDownload slide GPR records and electrical image for profile 1. 50 (a), 100 (b), 200 (c), and 400 MHz (d) GPR records, and electrical image (e). Major lithological changes in the near-subsurface are indicated. The shallowest reflector can be seen at the middle of the 400 MHz section at about 15 ns, which translates to an approximate depth of 0.75 m. Coring indicates that this reflector represents the base of the sandy soil, which grades into uncemented sand. Diffraction from a subsurface drainage pipe is apparent on the higher frequency GPR sections at around position 17 m at 20 ns TWT (around 1 m deep). The circular intermediate resistivity (<300 Ω m) patch at the middle of the electrical image in the near-surface may indicate this pipe. Weak diffraction patterns from the bouldery sediments in the soil zone can also be observed on the higher frequency sections. The reflection continuity has been disrupted by shallow diffractions, lithology changes, and interference from a direct ground wave. On the 100 MHz section, a reflector can be seen from the left part of the profile (at 35 ns time) to its right end (at approximately 25 ns in time). The 50 MHz record shows a strong reflector at 75 ns from the middle of the profile to its right end. The former reflector and the latter one seem to correlate with the top and bottom, respectively, of a coarser grained sandstone block. The shallow reflection is positive, while the deep one is negative. These reflections result from the changing moisture content at the surfaces of the coarser grained block. On the electrical image, this block can be identified as a high resistivity (>600 Ω m) patch. Repeat surveys (Berry 1996) have shown that this patch has been consistent throughout the surveys conducted over one month in June to July, 1996. The depths to any of the surfaces of the coarser grained sand block obtained from GPR and electrical imaging agree reasonably well with each other. The 100 MHz section does not show the base of the sandstone block, but a weak reflection can be seen at 45 ns from the middle of the profile to its right end. This may represent some change in lithology within the block. This block appears to extend throughout the entire right-hand part of the profile. In the left-hand part of the image line, there is a very low resistivity (<75 Ω m) patch at around 1.3 m depth. The tensiometer measurements indicate an unusually high moisture content at this depth (figure 4), which may be related to an abrupt decrease in grain size of the sediments. The GPR signal penetration depth abruptly decreases within this portion of the line, indicating the increased conductivity of the sediments in the shallow subsurface consistent with the above-mentioned interpretation. Figure 4. Open in new tabDownload slide Moisture content profile for the study area. MVF is the moisture volume fraction. Figure 4. Open in new tabDownload slide Moisture content profile for the study area. MVF is the moisture volume fraction. The depth of penetration of GPR signals varies from around 3 m at the left end of the profile to a maximum of around 7 m at position 35 m. The moderate penetration depth seems to be due to a high clay content in the shallow subsurface (1–3 m depth) and a decrease in the grain size of the sediments. If the clay content of the sediments increases while the grain size remains unchanged, the attenuation of the GPR signals may increase. From position 25 m onward, the GPR depth shows a local increase. As mentioned earlier, this is related to the presence of a relatively high resistivity block, possibly of coarser sands. The resistivity profile reveals an overall decrease in resistivity with depth, as confirmed from measuring the same line with double electrode spacing (and, so, with double length). The GPR surveys showed significant changes in the reflection pattern over short horizontal distances. The gradual decrease in the signal penetration depth crosses some sedimentary structures, indicating that it is linked to an abrupt change in lithology. Thus, the resistivity distribution and change in the GPR signal penetration depth indicate significant (vertical and lateral) heterogeneity of lithology and change in moisture content along the profile. The geophysical measurements of profile 2 are shown in figure 5. The GPR profiles of the subsurface sediments show almost the same features as revealed for profile 1. Reflection from the base of the sandy soil, however, seems to be absent or very weak on this profile. The GPR signal penetration depth varies from around 3 m at the left end of the profile to a maximum of around 6.5 m at position 32 m. The resistivity profile shows the same features as shown by profile 1. Figure 5. Open in new tabDownload slide GPR records and the electrical image for profile 2. 50 (a), 100 (b), 200 (c) and 400 MHz (d) GPR records, and the electrical image (e). Figure 5. Open in new tabDownload slide GPR records and the electrical image for profile 2. 50 (a), 100 (b), 200 (c) and 400 MHz (d) GPR records, and the electrical image (e). Profile 3 is characterized by the highest attenuation of the GPR signals among the three profiles studied (figure 6). Any reflection from the base of the coarser grained sandstone block is absent from the records. The radar penetration depth is moderate, and is almost constant throughout the profile: around 3.5 m and 4.5 m in the left-hand end and right-hand end of it, respectively. The almost total loss of energy may be due to a higher clay content of the sand, and a decrease in grain size within this part of the study area. The resistivity image fully supports this interpretation. Figure 6. Open in new tabDownload slide GPR records and the electrical image for profile 3. 50 (a), 100 (b), 200 (c) and 400 MHz (d) GPR records, and the electrical image (e). Figure 6. Open in new tabDownload slide GPR records and the electrical image for profile 3. 50 (a), 100 (b), 200 (c) and 400 MHz (d) GPR records, and the electrical image (e). The subsurface features can usually be traced from one line to the next, confirming that they are real and not caused by instrumental or other problems, and appear to run almost parallel to the topographic dip direction. The results of GPR and electrical profilings indicate the capability of these tools to identify the variation in lithology and in moisture content. The results of these observations indicate the higher degree of heterogeneity of the top Triassic sandstones. The continuity of reflections was not always observed due to interference among the reflections, and from shallow diffractions and a direct ground wave, and lateral changes in the signal penetration depth. A contour map of the GPR signal penetration depth (for 50 MHz) is shown in figure 7. It reveals a relatively deeper penetration, reaching the value of 7 m in the right-hand part of the study area. As mentioned earlier, this deeper penetration can be attributed to a coarser grained sandstone block. The reduced penetration (around 3 m) observed in the left-hand part of the study area can be correlated with a decrease in grain size and increased clay content. The occurrence of an increased amount of fine-grained sediments and clay content at depths below 5 m is confirmed by a high moisture content (from long tube moisture content profile) (figure 8). Figure 7. Open in new tabDownload slide Penetration depth of GPR signals (m). The contour interval is 0.2 m. Figure 7. Open in new tabDownload slide Penetration depth of GPR signals (m). The contour interval is 0.2 m. Figure 8. Open in new tabDownload slide Long tube moisture content profile. MVF is the moisture volume fraction. Figure 8. Open in new tabDownload slide Long tube moisture content profile. MVF is the moisture volume fraction. 5. Discussion and conclusions At the present study site, variations in lithology and moisture content, and some small-scale structures, may have been identified within the top of UK Triassic sandstones using high-resolution electrical imaging and GPR profiling. Strong reflections have been observed from the upper and lower surfaces of an assumed coarser sand block. This block has also been reflected in electrical images. The GPR profiling and electrical resistivity imaging reveal a higher degree of spatial (vertical and lateral) heterogeneity of the high-conductivity sandstones, implying spatial variations in moisture content. Since the propagation velocity of radar waves is largely determined by the soil water content (Davis and Annan 1989), lateral radar-wave velocity variations could be expected. Thus, the lateral variations in moisture content can distort radar images of the subsurface, e.g., affecting the continuity and clarity of reflections. The comparison of the GPR and electrical images from adjacent lines provides the possibility of visualizing the three-dimensional aspects of the subsurface structures. The radar signal penetration depth ranges from about 3 to 7 m at this site. The reduced penetration depth seems to be due to a high clay content in the shallow subsurface, and a decrease in the grain size of the sediments. The field results reveal the problem of differentiating grain size variation from the variation in the clay content of the sand. Without very good information about the sedimentology, it is difficult to say which of the two factors, grain size or clay content, influences the electrical properties in a particular case. GPR profiling and electrical imaging are rapid, economical and reliable methods for high resolution and continuous mapping of the subsurface sediments. For investigating conductive sandy sediments, like the UK Triassic sandstones, using a combination of radar antenna frequencies (starting from as low as possible) and electrical imaging with small unit electrode spacing (fractions of a m) seems to be more appropriate. Continuous borehole logs could be useful in the accurate interpretation of the above-mentioned data. Acknowledgments The author expresses his thanks to the Association of Commonwealth Universities for sponsoring this work. The author extends special thanks to Dr J H Tellam and Dr R D Barker of the School of Geography, Earth and Environmental Sciences at the University of Birmingham for research and logistical support. Roger Livesey of the same School is gratefully acknowledged for field assistance. Prof. Dr A S M Woobaid Ullah, Department of Geology, Dhaka University, is thanked for his suggestions. The author sincerely thanks two anonymous referees for critically reviewing the manuscript and for useful suggestions. References Andrews R J , Barker R D , Loke M H . , 1995 The application of electrical tomography in the study of the unsaturated zone in chalk at three sites in Cambridgeshire, United Kingdom , Hydrogeology J. , vol. 3 (pg. 17 - 31 ) 10.1007/s100400050055 Google Scholar Crossref Search ADS WorldCat Crossref Arora T , Ahmed S . , 2011 Characterization of recharge through complex vadose zone of a granitic aquifer by time-lapse electrical resistivity tomography , J. Appl. Geophys. , vol. 73 (pg. 35 - 44 ) 10.1016/j.jappgeo.2010.11.003 Google Scholar Crossref Search ADS WorldCat Crossref Barker R D . , 1996 Recent applications of electrical imaging surveys in the United Kingdom Proc. SAGEEP (pg. 1289 - 1302 ) Beres M Jr , Haeni E P . , 1991 Application of ground-penetrating-radar methods in hydrogeologic studies , Ground Water , vol. 29 (pg. 375 - 386 ) 10.1111/j.1745-6584.1991.tb00528.x Google Scholar Crossref Search ADS WorldCat Crossref Berry K . , 1996 Electrical imaging of water saturation changes in the unsaturated zone during pumping tests , MSc Thesis University of Birmingham Chambers J E , Kuras O , Meldrum P I , Ogilvy R D , Hollands J . , 2006 Electrical resistivity tomography applied to geologic, hydrogeologic, and engineering investigations at a former waste-disposal site , Geophysics , vol. 71 (pg. B231 - B239 ) 10.1190/1.2360184 Google Scholar Crossref Search ADS WorldCat Crossref Davis J L , Annan A P . , 1989 Ground penetrating radar for high-resolution mapping of soil and rock stratigraphy , Geophys. Prospect. , vol. 37 (pg. 531 - 551 ) 10.1111/j.1365-2478.1989.tb02221.x Google Scholar Crossref Search ADS WorldCat Crossref de Groot-Hedlin C , Constable S . , 1990 Occam's inversion to generate smooth two-dimensional models from magnetotelluric data , Geophysics , vol. 55 (pg. 1613 - 1624 ) 10.1190/1.1442813 Google Scholar Crossref Search ADS WorldCat Crossref Digges La Touche S V . , 1998 Unsaturated flow in the Triassic sandstones of the UK , PhD Thesis University of Birmingham Fisher E , McMechan G A , Annan A P . , 1992 Acquisition and processing of wide-aperture ground penetrating radar data , Geophysics , vol. 57 (pg. 495 - 504 ) 10.1190/1.1443265 Google Scholar Crossref Search ADS WorldCat Crossref Geotomo Software . , 2012 Rapid 2D resistivity and IP inversion using the least-squares method RES2DINVx32 version 3.71 www.geotomosoft.com Gerlitz K , Knoll M D , Cross G M , Luzitano R D , Knight R . , 1993 Processing ground penetrating radar data to improve resolution of near-surface targets Proc. SAGEEP (pg. 561 - 574 ) Giao P H , Weller A , Hiera D H , Adisornsupawat K . , 2008 An approach to construct the weathering profile in a hilly granitic terrain based on electrical imaging , J. Appl. Geophys. , vol. 65 (pg. 30 - 38 ) 10.1016/j.jappgeo.2008.03.004 Google Scholar Crossref Search ADS WorldCat Crossref Gómez-Ortiz D , Martín-Crespo T , Rodríguez I , Sánchez M J , Montoya I . , 2009 The internal structure of modern Barchan dunes of the Ebro river delta (Spain) from ground penetrating radar , J. Appl. Geophys. , vol. 68 (pg. 159 - 170 ) 10.1016/j.jappgeo.2008.11.007 Google Scholar Crossref Search ADS WorldCat Crossref Interpex Ltd . , 1996 , Gradix VI: Ground Penetrating Radar Processing and Interpretation Golden, CO Interpex Ltd Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Jol H M . , 2009 , Ground Penetrating Radar: Theory and Applications Amsterdam Elsevier Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Loke M H , Barker R D . , 1995 Least-squares deconvolution of apparent resistivity pseudosections , Geophysics , vol. 60 (pg. 1682 - 1690 ) 10.1190/1.1443900 Google Scholar Crossref Search ADS WorldCat Crossref Loke M H , Barker R D . , 1996 Rapid least-squares inversion of apparent resistivity pseudosections by a quasi-Newton method , Geophys. Prospect. , vol. 44 (pg. 131 - 152 ) 10.1111/j.1365-2478.1996.tb00142.x Google Scholar Crossref Search ADS WorldCat Crossref MALA . , 2012 www.malags.com Pellicer X M , Gibson P . , 2011 Electrical resistivity and ground penetrating radar for the characterisation of the internal architecture of Quaternary sediments in the midlands of Ireland , J. Appl. Geophys. , vol. 75 (pg. 638 - 647 ) 10.1016/j.jappgeo.2011.09.019 Google Scholar Crossref Search ADS WorldCat Crossref Pipan M , Forte E , Dal Moro G , Sugan M , Finetti I . , 2003 Multifold ground-penetrating radar and resistivity to study the stratigraphy of shallow unconsolidated sediments , The Leading Edge , vol. 22 9 (pg. 876 - 881 ) 10.1190/1.1614161 Google Scholar Crossref Search ADS WorldCat Crossref Turesson A . , 2006 Water content and porosity estimated from ground-penetrating radar and resistivity , J. Appl. Geophys. , vol. 58 (pg. 99 - 111 ) 10.1016/j.jappgeo.2005.04.004 Google Scholar Crossref Search ADS WorldCat Crossref Turner G , Ingham M , Bibby H , Keys H . , 2011 Resistivity monitoring of the tephra barrier at Crater Lake, Mount Ruapehu, New Zealand , J. Appl. Geophys. , vol. 73 (pg. 243 - 250 ) 10.1016/j.jappgeo.2011.01.006 Google Scholar Crossref Search ADS WorldCat Crossref Vaudelet P , et al. , 2011 Mapping of contaminant plumes with geoelectrical methods. A case study in urban context , J. Appl. Geophys. , vol. 75 (pg. 738 - 751 ) 10.1016/j.jappgeo.2011.09.023 Google Scholar Crossref Search ADS WorldCat Crossref © 2013 Sinopec Geophysical Research Institute
TI - Evaluation of ground penetrating radar and resistivity profilings for characterizing lithology and moisture content changes: a case study of the high-conductivity United Kingdom Triassic sandstones
JF - Journal of Geophysics and Engineering
DO - 10.1088/1742-2132/10/6/065003
DA - 2013-12-01
UR - https://www.deepdyve.com/lp/oxford-university-press/evaluation-of-ground-penetrating-radar-and-resistivity-profilings-for-uXXRLG6tV7
VL - 10
IS - 6
DP - DeepDyve
ER -